LLM Refusal Behavior

LLM: Large Language Model

The field asks when an LLM should decline, whether it can distinguish harmful requests from benign or dual-use ones, and how refusals affect users and agents. Researchers use matched benchmarks and audits to measure harmful compliance, over-refusal, and partial compliance across settings such as health, biology, and cybersecurity, including varied prompts, multi-turn conversations, tools, and sampling conditions. Mechanistic studies use activation directions, probes, sparse autoencoders, and causal tracing to investigate refusal signals, while steering, decoding controls, and fine-tuning aim to adjust refusal behavior. A central tension is that reducing harmful compliance can increase unnecessary refusals, and interventions may have side effects, so a single refusal rate is not enough to establish safe, useful behavior.

Latest papers

67 papers7 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

Measuring and Eliminating Refusals in Military Large Language Models
Feb 18, 2026Jack FitzGerald, Dylan Bates, Aristotelis Lazaridis +17Refusal Behavior in Language Models LLM Refusal Behavior
EdgeRunner AI.
A Content-Based Framework for Cybersecurity Refusal Decisions in Large Language Models
Feb 17, 2026Noa Linder, Meirav Segal, Omer Antverg +5LLM Refusal Behavior Refusal Behavior in Language Models
Irregular · University of Zurich
$C$ - $ΔΘ$ : Circuit-Restricted Weight Arithmetic for Selective Refusal
Feb 4, 2026Aditya Kasliwal, Pratinav Seth, Vinay Kumar SankarapuLLM Refusal Behavior Causal Interventions in Language Models
Lexsi Labs
There Is More to Refusal in Large Language Models than a Single Direction
Feb 2, 2026Faaiz Joad, Majd Hawasly, Sabri Boughorbel +2LLM Interpretability LLM Refusal Behavior
Qatar Computing Research Institute, HBKU, Doha, Qatar
Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health Context
Jan 25, 2026Zhihao Zhang, Liting Huang, Guanghao Wu +3LLM Safety Benchmarks Language Model Safety Evaluation
Macquarie University1 · UTS2 · MBZUAI3 +1
The Instability of Safety: How Random Seeds and Temperature Expose Inconsistent LLM Refusal Behavior
Dec 12, 2025Erik LarsenLanguage Model Safety Evaluation LLM Refusal Behavior
LLMs Encode Harmfulness and Refusal Separately
Jul 16, 2025Jiachen Zhao, Jing Huang, Zhengxuan Wu +2LLM Interpretability LLM Refusal Behavior
Northeastern University · Stanford University

Feb 18, 2026J/K move · Enter open · S save

Measuring and Eliminating Refusals in Military Large Language Models

Jack FitzGerald, Dylan Bates, Aristotelis Lazaridis, Aman Sharma+16

EdgeRunner AI.

Refusal Behavior in Language Models LLM Refusal Behavior LLM Safety Benchmarks

Open paper AskPDF arXiv

Military Large Language Models (LLMs) must provide accurate information to the warfighter in time-critical and dangerous situations. However, today's LLMs are imbued with safety behaviors that cause the LLM to refuse many legitimate queries in the military domain, particularly those related to violence, terrorism, or military technology. Our gold benchmark for assessing refusal rates, which was developed by veterans of the US Army and special forces, is to our knowledge the first dataset of its kind. We present results for refusal and deflection rates on 31 public models and 3 military models. We observe hard rejection rates as high as 98.2% and soft deflection rates ranging from 0% to 21.3%. We also present results on two additional synthetic datasets and show their correlations with the gold dataset. Finally, we perform abliteration using the Heretic library on a military-tuned gpt-oss-20b model, showing an absolute increase in answer rate of 66.5 points but an average relative decrease of 2% on other military tasks. In our concluding remarks, we argue for deeper specialization, including with mid-training and end-to-end post-training, to achieve zero refusals and maximum military task accuracy for closed military models.